An application on parametric survival analysis and kidney graft

Survival analysis deals with the expected expected failure time of a mechanical or electronic component. It is a branch of statistics that deals with the expected time of death of a living thing, and it is used in socio-economic fields as well as closely related to the theory of reliability in engineers. In the survival analysis, the expected time estimation is modeled by making various assumptions. In the event of deterioration, failure, or death in the survival analysis, the event has occurred. Survival analysis is also used in patient and disease follow-up. After the diagnosis is made, the patient is followed up and the survival time is examined. After the treatment is over, estimates are made about how long he will survive and the recurrence of the disease. The variables affecting the survival time are examined. In this thesis, survival analysis, survival time, censored data and survival time functions are explained and continues with parametric survival distributions. Other parametric survival distributions such as Cox Proportional Hazards model, Weibull, Exponential, Rayleigh and Log-normal were investigated. Then, the python coding language, which directly models the survival function instead of estimating the Hazard Function to predict the survival time of kidney graft patients, shows that it predicts how many of the patients survive with the accuracy value. İn addition, survival analysis was performed with Kaplan Meier in SPSS and compatibility with distributions was examined in Minitab.